Abstract Odontogenic neoplasms, including ameloblastomas and keratocystic odontogenic tumors, demonstrate locally aggressive and destructive behavior. A variety of surgical treatments may be used for these neoplasms; however, their location makes complete resection difficult, leading to residual disease and high rates of recurrence. While over-aggressive resection may be carried out to reduce residual disease, this results in greater disfiguration and reduced facial function for the patient. Our work on these tumors has focused on understanding their genetics, examining new therapies, and developing new models. At present, there are currently no biomarkers used for these tumor types, and no diagnostic strategies beyond standard biopsy. The ability to accurately assess tumor margins with specific, non-invasive imaging could result in the preservation of healthy tissue and improvements in long-term local tumor control. We hypothesize that epidermal growth factor receptor (EGFR) expression in aggressive odontogenic neoplasms may be used to specifically visualize tumors intraosseously, which may allow the tumor margins to be assessed intraoperatively. Our approach utilizes FDA-approved fluorescent and radiolabeled epidermal growth factor (EGFR) antibodies and imaging technology. We aim to establish the clinical validity of using panitumumab as an imaging agent for ameloblastomas. The long-term goal of our studies is to implement non-invasive biomarker-based imaging of ameloblastomas in clinical care, making it possible to precisely label tumor tissue and assess tumor margins either pre- or intraoperatively, allowing clinicians to provide better care for their patients. The objective of our current studies is to determine the sensitivity and specificity of labeled anti-EGFR antibodies to image ameloblastoma tissue using in vivo preclinical models. Two hypothesis-driven specific aims will be investigated as follows: (1) To determine the in vivo sensitivity and specificity of panitumumab-IRDye800 for human ameloblastoma patient-derived xenografts (PDX). We hypothesize that panitumumab-IRDye800 will localize to ameloblastoma tumor tissue significantly more than a non-specific IgG-IRDye800 antibody and that the fluorescence will be higher than the background fluorescence of normal tissue. (2) To determine the clinical validity of panitumumab-IRDye800 and 89Zr-panitumumab for the surgical removal of tumors using an intraosseous mandibular model of ameloblastoma. We hypothesize that both panitumumab-IRDye800 and 89Zr- panitumumab will localize to ameloblastoma cells and allow accurate margin determination and surgical removal of tumors. This project has the potential to develop a method to accurately image ameloblastomas and provide a tool for assessing bone invasion in a patient population that is vastly under-represented in the existing research. During the K99 and R00 portions of this grant, we have created patient-derived xenograft mouse models of ameloblastoma and are continuing patient recruitment to develop new intraosseous models. The imaging strategies that are being developed based on our findings will make it possible to non-invasively and accurately image tumors. This can help determine the area of resection in order to obtain clear margins, while also reducing the resection of healthy tissue. Thus, this research has the potential to directly impact patient care.